2014 Lincoln MKTEnlarge Photo
For the most part, cruise control can save a lot of fuel—up to 10 percent, versus foot-on-accelerator driving—by smoothing our driving styles. Yet the more we 'adapt' cruise control to surrounding traffic or more extreme terrain, the more we need to question the way in which it pulls off the basics—maintaining speed, and 'recovering' back to a desired speed when vehicles (or hills) aren't slowing us down.
Virtually all the cruise control systems now on the market (even so-called adaptive cruise controls that set a following distance) are not predictive. They merely push harder on the throttle when speed drops or there's no longer a slower vehicle directly ahead, and ease off when speed rises back to the desired setting.
That's led engineers at Ford to rethink cruise-control behavior. The idea is to eliminate some of the overcompensation that's built into these systems.
With some predictive algorithms and more sophisticated mapping, the Ford engineers hope to soften throttle behavior—and save more fuel—when possible.
Powertrain mapping for Ford predictive cruise controlMapping it out
Powertrain mapping for Ford predictive cruise controlEnlarge Photo
That means mapping each vehicle's powertrain in greater detail, the engineers explained, and comparing the load on the vehicle (how many people are aboard, or whether there's a headwind), and the steepness of a grade, to the powertrain torque that can and should be applied, for a smooth 'recovery trajectory' if the speed drops.
Fully connected and cloud-based vehicles might know what's up ahead, but based on a vehicle's onboard measurements on powertrain load, accelerometers, and yaw sensors—and the air-mass meter, and any of the numbers that are in a typical powertrain module—the system can make a pretty reasonable guess.
The system looks at those variables, in samples of a few seconds at a time, with a total running memory of less than 55 seconds, and predicts based on recent loading and terrain whether the hill will level off soon, or whether the load is a constant or transient one (one that it thinks might let up soon).
Setting the stage for connected vehicles
It's a clever solution that requires no additional hardware, and the Dynamic Cruise Control that results could be used as a base line for more effective adaptive cruise control, active-safety systems, and in the longer term, autonomous-driving situations.
So if you've gone over a series of small, undulating hills every quarter of a mile or so, the system might 'guess' that there's no need to aggressively increase the throttle, and it will instead wait until the next time it levels off—or even use the downhill.
That makes it quite different than the so-called 'eco' cruise-control modes offered in some vehicles—which merely put a rubber-band-style delay on the throttle response.